Grok 4.7 Xhigh vs Qwen3.8 Max 0902 compared

Grok 4.7 (Xhigh) from SpaceXAI and Qwen3.8 Max (0902) from Alibaba land within 1.8 points of each other on the intelligence index. No hands-on testing went into this write-up; it summarises the recorded benchmark, price and speed data for both models. What separates them most is responsiveness; Grok 4.7 Xhigh takes 43 seconds to produce its first token, Qwen3.8 Max 0902 3.2 seconds.

Grok 4.7 Xhigh is the faster writer

Measured output speed is 65 tokens per second for Grok 4.7 Xhigh and 38 tokens per second for Qwen3.8 Max 0902. Grok 4.7 Xhigh is ahead by 70% on throughput. Before any text appears, Grok 4.7 Xhigh takes 43 seconds and Qwen3.8 Max 0902 3.2 seconds. So the two pull in opposite directions, with Qwen3.8 Max 0902 quicker off the mark and Grok 4.7 Xhigh quicker once it gets going; the crossover is around 3,700 tokens. Waits that long are felt on every chat-style request, though they matter less for batch jobs.

For a 1,000-token response, first token plus generation comes to roughly 59 seconds on Grok 4.7 Xhigh and 29 seconds on Qwen3.8 Max 0902.

Pricing: no difference

Per million tokens, Grok 4.7 Xhigh charges $2.00 for input and $6.00 for output, while Qwen3.8 Max 0902 charges $2.00 and $6.00. The rates are identical, so price will not separate them.

Which is newer

Grok 4.7 Xhigh arrived on September 21, 2026, 19 days after Qwen3.8 Max 0902 (September 2, 2026). They are effectively contemporaries.

Overall capability: 1.8 points apart

On the intelligence index, Grok 4.7 Xhigh scores 35.5 against 33.7 for Qwen3.8 Max 0902. It is a small margin for an index that averages many evaluations. Put another way, Qwen3.8 Max 0902 would need 5% more on this index to draw level.

Among 109 active scored models, Grok 4.7 Xhigh sits 38th and Qwen3.8 Max 0902 44th.

Neither model has a separate coding or math index recorded yet, so those narrower comparisons will have to wait for more data.

Score per dollar

Per point of intelligence index, Grok 4.7 Xhigh costs $0.085 of blended price per million tokens and Qwen3.8 Max 0902 $0.089. So neither is a bargain relative to the other; you pay for what the index measures.